Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

ukr-gpt #4

Open
wants to merge 32 commits into
base: adaptation
Choose a base branch
from
Open

ukr-gpt #4

wants to merge 32 commits into from

Conversation

koren-v
Copy link

@koren-v koren-v commented Aug 28, 2021

No description provided.

Paul Khudan and others added 30 commits May 31, 2021 15:50
refactor pipeline. resolve task <-> sandbox relationship
d070d2e 
# Conflicts:
#	src/language_model/data/extract.py
#	src/language_model/data/load.py
#	src/language_model/modelling/trainer.py
#	src/language_model/tokenization/extend.py
#	src/language_model/tokenization/tasks.py
#	src/language_model/tokenization/trainer.py
combine two logger-initting methods
e12a87e
Merge pull request #1 from youscan/infra
80187b2 
Infra
@dchaplinsky
Update README.md
e1388b5
@dchaplinsky
Merge pull request #1 from lang-uk/dchaplinsky-patch-1
a495a50 
Making an example to work
@terpiljenya
Merge pull request #3 from lang-uk/master
3670dac 
Making an example to work.
@koren-v
Merge remote-tracking branch 'origin/adaptation' into gpt
f01b9d7 
# Conflicts:
#	src/language_model/modelling/trainer.py
#	src/language_model/runner.py
#	src/language_model/tokenization/trainer.py
@koren-v
ukr-gpt, FromIterableTextDataset, GroupTextForCasualLMDataset
8c4adf9
@koren-v
MinHashLSH deduplication, Wiki
fd92a6b
@koren-v
MinHashLSH deduplication, Wiki
e4ae992
@koren-v
configs
e41d5b1
@koren-v
updated with experiments
3513970
@koren-v
saving PreTrainedTokenizer
1e1f1f6
@koren-v
new configs
4300cfd
@koren-v
Iterable[Iterable[str]] -> Iterable[str] (should be merged outside)
e889a3f
@koren-v
writing to single/train-val files
acf89ed
@koren-v
steps
6d56975
@koren-v
pre-commit fixes
d6f5d73
@koren-v
change paths, save_total_limit, comment lsh dependency
13af9e1
@koren-v
separate data step: saving input ids
66b25c8
@koren-v
separate MinHashLSHDeduplicator
ae77dae
@koren-v
convert tokenizer to transformers format
5d0566e
@koren-v
removed unused class
2b4103a
@koren-v
updated config
7a34448
@koren-v
map -> submit & as_completed
ae0a8c9
@koren-v
IterableDataset
0b1f91d
@koren-v
pre-commit fixes
5bd1dc0
@koren-v
update paths
eb8aac7
@koren-v
fix max length
c3fdeb5
@koren-v
more frequent eval
ba2569c
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@koren-v @dchaplinsky @terpiljenya